Arto Suvas’s research while affiliated with University of Vaasa and other places

This study compares the accuracy and efficiency of five different estimation methods for predicting financial distress of small and medium-sized enterprises. We apply different methods for a large set of financial and non-financial variables, using filter and wrapper selection, to predict bankruptcy up to 10 years before the event in an open, European economy. Our findings show that logistic regression and neural networks are superior to other approaches. We document how the cost-return ratio considerably affects the location of optimal cut-off points and attainable profit in credit decisions. Once a loan provider selects a particular prediction model, an effort should be made to find the optimal cut-off score to maximize the efficiency of the technique. Indeed, this often involves determining several cut-off levels where the portfolio of products and services exhibits different cost-return characteristics.

This paper sheds some new light on banks’ distress in Europe, with special attention paid to the period of the global financial crisis (GFC). Unlike in previous research we investigate non-distress (“glory”) and distress (“shame”) paths of banks from 1 to 4 years prior to a distress event to test how different they are. This approach allows us to outline guidelines for supervisors on how to detect banks generating higher risk of distress several years before its occurrence. We use a balanced panel of data, applying factor and cluster analysis for extraction of distress processes and a logistic regression for distress prediction. We conclude that the differences between distressed and non--distressed banks become more visible 1 and 2 years prior to the distress event. However, liquid assets and loans to assets ratios are significant and stable predictors of banks’ distress even 3−4 years in advance.

This paper assesses the classification performance of the Z-Score model in predicting bankruptcy and other types of firm distress, with the goal of examining the model's usefulness for all parties, especially banks that operate internationally and need to assess the failure risk of firms. We analyze the performance of the Z-Score model for firms from 31 European and three non-European countries using different modifications of the original model. This study is the first to offer such a comprehensive international analysis. Except for the United States and China, the firms in the sample are primarily private, and include non-financial companies across all industrial sectors. We use the original Z''-Score model developed by Altman, Corporate Financial Distress: A Complete Guide to Predicting, Avoiding, and Dealing with Bankruptcy (1983) for private and public manufacturing and non-manufacturing firms. While there is some evidence that Z-Score models of bankruptcy prediction have been outperformed by competing market-based or hazard models, in other studies, Z-Score models perform very well. Without a comprehensive international comparison, however, the results of competing models are difficult to generalize. This study offers evidence that the general Z-Score model works reasonably well for most countries (the prediction accuracy is approximately 0.75) and classification accuracy can be improved further (above 0.90) by using country-specific estimation that incorporates additional variables.

Reviews of financial distress prediction models indicate that these techniques give highly reliable estimates of probabilities of default only for relatively short horizons, rarely exceeding two years. This is particularly the case when financial variables make up the sole estimate or primary estimates. So far, bank managers have focused on the one-year probability of default estimation required by Basel capital regulations. According to an emerging accounting standard (IFRS 9), banks will be obligated to estimate a probability of default lifetime in order to calculate credit allowances. Moreover, there is a need to improve communication and transparency between small and medium-sized private enterprises and suppliers of capital to overcome the problem of credit rationing, especially in Europe. Thus, it is challenging to search for newtools to extend the distress or failure prediction period. We assess the long-term (up to ten years) predictive ability of both financial and nonfinancial variables, paying special attention to the role of nonfinancial variables. Our study is based on rigorous postdevelopment distress and nondistress financial events in the Finnish environment. Our model, built with cross-sectional data from 2003, analyzes results for 2004-13 and shows that measures of solvency, turnover, environmental risk, payment behavior and board member characteristics can be significant predictors of bankruptcies for as long as ten years. Our most accurate long-range prediction results combine financial and nonfinancial variables.

Purpose The purpose of this paper is to find out which different failure processes exist among the young manufacturing micro firms, and whether the representation of those processes differs first, in European countries, and second, among exporting and non-exporting firms. Design/methodology/approach The study is based on financial data of 1,216 manufacturing micro firms from European countries. Failure processes have been detected with a two stage-method: by extracting latent dimensions from financial variables with factor analysis, and then, by clustering the established factor scores. Findings With firms’ age, the number of different failure processes reduces from four to two. Strong evidence was found about the dominance of different failure processes in different countries for most firm age groups. Failure processes are not strongly associated with (non-)exporting. Originality/value This paper is the first one determining young manufacturing micro firms’ failure processes and comparing the representation of those processes in different firm subsets, either based on their country of origin or (non-)exporting behavior. Moreover, previous studies have not encompassed specific sectors, young or very small firms.

The objective is to investigate the influence of Hofstede’s original cultural dimensions on financial distress prediction. Empirical data consist of 1,255,768 non-failed and 22,594 failed yearly firm observations from 26 European countries. First, an overall six variable logistic regression model is estimated to predict financial distress in an international context. Second, logistic regression models including moderating (interaction) effects with each financial predictor variable are separately estimated for each cultural dimension. Empirical findings show that Hofstede’s dimensions significantly moderate the effects of many financial predictors in failure prediction. However, equity ratio (solvency) and return on investment ratio (profitability) play central roles in prediction models irrespective of moderating effects. Therefore, solvency and profitability are useful predictors of financial distress in international modeling. Due to the dominant role of the equity ratio across cultures, the contributions of moderating effects and further variables on the overall performance of prediction models are not strong.

Financial distress prediction models developed by researchers in different countries differ significantly from each other with respect to the variables used and the coefficients estimated. Therefore, the use of country-specific models in international analysis across countries is questionable. National culture is an important factor behind the differences in the estimated models. The objective of this study is to investigate how national culture, in terms of Hofstede's cultural dimensions, affects the coefficients of country-specific models. A set of six financial variables is selected on theoretical grounds, and a financial distress prediction model is estimated by the logistic regression analysis for a group of European countries, separately for each country. The data include financial variables of 3,372,493 non-failed and 56,541 failed firm observations from 30 European countries. Then, the impact of cultural dimensions on the coefficients of the models is assessed by correlation and regression analyse...

This study focuses on the pre-failure growth in total assets, debt and sales of bankrupted manufacturing firms. Based on a sample of 128 Estonian firms, it is shown that two distinct growth patterns can be outlined. When the first pattern shows a gradual decline, then the other characterizes a more eclectic growth behavior. Several classical financial ratios have significantly different values through the established two patterns. Managers' characteristics do not vary among the established patterns.

Reviews on financial distress prediction models indicate that these techniques give highly reliable estimates of probabilities of default (PDs) and loss given default (LGD) only for relatively short horizons, rarely beyond two years. Major stakeholders, e.g. investors and bank risk and capital analysts, therefore, have such models sanctioned by portfolio managers and regulators for the same short horizons; for example, the Basel Committee on Banking Supervision recommends PD and LGD estimates for one year. This is especially the case when financial variables make up the sole or primary estimates, and only a bit longer reliable estimators when these models include non-financial variables as additional early warning signals. Beyond three years, such models, regardless of their structure, rarely give reliable estimates, perhaps not much better than flipping a coin. The objective of this study is to assess the predictive ability of both financial and non-financial variable constructs for longer term horizons of up to ten years based on rigorous post-development distress and non-distress financial events in the Finnish environment. Our model, built with cross-section data from 2003, analyses results for 2004-2013. Results show that measures of solvency, turnover, industry risk, payment behaviour, and board member characteristics can be significant predictors of bankruptcies for as long as ten years. The most accurate long-range prediction results combine financial and non-financial variables. Subsequent tests should attempt to extend such models in a multi-country setting, whether or not bankruptcy regimes are similar across national borders.